HHS Public Access Author manuscript Author Manuscript
Gynecol Oncol. Author manuscript; available in PMC 2017 August 01. Published in final edited form as: Gynecol Oncol. 2016 August ; 142(2): 299–303. doi:10.1016/j.ygyno.2016.06.009.
Who Presents Satisfied? Non-modifiable Factors Associated with Patient Satisfaction among Gynecologic Oncology Clinic Patients Emma L Barber, MD1,2, Jeannette T Bensen, PhD2,3, Anna Snavely, PhD4, Paola A Gehrig, MD1,2, and Kemi M Doll, MD1,2,5
Author Manuscript
1Division
of Gynecologic Oncology, Department of Obstetrics and Gynecology, University of North Carolina, Chapel Hill, NC
2Lineberger
Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC
3Department
of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC
4PDstat
LLC
5Department
of Health Policy and Management, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC
Abstract Author Manuscript
Objective—To examine associations between non-modifiable patient factors and patient satisfaction (PS) among women presenting to a gynecologic oncology clinic. Methods—This is a cross sectional analysis of patients presenting for surgical management by a gynecologic oncologist at a tertiary care academic medical center. The Patient Satisfaction Questionnaire (PSQ-18) that measures PS in seven domains of health care was administered. Scores were converted to “satisfied” versus “unsatisfied/equivocal”. Demographic and medical factors were obtained from the medical record. Chi-square, t-tests, and multivariable logistic regression were used.
Author Manuscript
Results—208 patients completed the baseline patient satisfaction questionnaire and the median PSQ-18 score was 70.5 (range 42–90). Median age was 58 years (range 22–93). Several non-modifiable factors were associated with PS. White patients had higher interpersonal PS than minorities (86% v 65% p =0.002). The uninsured had lower interpersonal (60% v 86% p =0.003)
CORRESPONDING AUTHOR: Emma Barber, MD, Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, University of North Carolina at Chapel Hill, 103B Physicians’ Office Building, Campus Box #7572, Chapel Hill, NC 27599. Phone: 919-966-1194; Fax: 919-843-5387; [email protected]. Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. PRESENTATIONS: The data contained within this article was presented in abstract form at both the Annual Winter Meeting of the Society of Gynecologic Oncology February, 2016 in Squaw Valley, CA and at the Society for Gynecologic Oncology Annual Meeting March, 2016 in San Diego, CA. CONFLICT OF INTEREST STATEMENT: The authors declare there are no conflicts of interest.
Barber et al.
Page 2
Author Manuscript
and accessibility PS (33% v 67% p =0.03). Increasing education and less time traveled to care were both associated with higher interpersonal (p =0.03, p =0.05) and accessibility PS (p =0.01, p =0.01). There was no association between clinical factors (BMI, comorbidities, cancer) and PS. In multivariable analysis, the strongest predictor of interpersonal PS was white race while the strongest predictors of accessibility PS were time travelled to care and insurance status. Conclusions—Patient satisfaction scores among patients presenting to a gynecologic oncology clinic are associated with non-modifiable demographic, financial and geographic factors. Pay for performance measures that use summed patient satisfaction scores may penalize hospitals for patient-mix driven differences.
INTRODUCTION Author Manuscript
The Affordable Care Act mandates that payments to hospitals depend on providing valuebased care (1). Two percent of total hospital Center for Medicaid and Medicare Services (CMS) reimbursements is calculated using value-based performance scores (2). Within these value-based performance scores, a subjective measure of performance called the ‘patient experience of care domain’, is 30% of the total hospital performance score (2). This domain is measured using a patient satisfaction questionnaire.
Author Manuscript
Given that patient satisfaction is part of how the delivery of value-based care is measured, many investigators have examined the relationship between patient satisfaction and other markers of high quality healthcare with mixed results. In some studies, higher patient satisfaction scores have been correlated with decreased hospital readmissions, decreased length of hospital stay and decreased mortality (3). Others studies, specifically in surgery, have found no relationship between patient satisfaction and adherence to Surgical Care Improvement Program (SCIP) process measures, postoperative complications or mortality (4, 5). Paradoxically, one study actually found an inverse relationship: patient satisfaction was associated with increased healthcare costs and increased mortality (6).
Author Manuscript
In addition to studying the relationship between patient satisfaction and other markers of high quality healthcare, investigators have also examined the association between nonmodifiable patient factors and patient satisfaction. In this work, patient satisfaction scores vary with a number of non-modifiable patient and hospital system factors. At the patient level, lower satisfaction scores have been found for younger and more educated patients (7, 8). Other studies have found lower satisfaction scores for patients who are non-English speakers (8). At the systems level, lower patient satisfaction scores have been observed for densely populated urban areas, hospitals with more beds and those that are non-profit or academic (8, 9). The consistent variation of patient satisfaction by non-modifiable factors suggests that patient satisfaction may be measuring factors beyond just the quality of healthcare delivered. However, much of this literature focuses on the inpatient setting, and the association between non-modifiable factors and patient satisfaction has not been studied to date in outpatient gynecologic oncology patients. Our objective was to evaluate the association between non-modifiable patient factors and domain-specific patient satisfaction scores among women presenting to a gynecologic oncology clinic. Gynecol Oncol. Author manuscript; available in PMC 2017 August 01.
Barber et al.
Page 3
Author Manuscript
METHODS
Author Manuscript
We conducted a cross sectional analysis of a larger parent prospective cohort study: The Impact of Surgical Complications on Health Related Quality of Life.(10) The recruitment of this cohort has been previously described. Briefly, patients were identified and enrolled in a hospital-based Registry, the University of North Carolina at Chapel Hill (UNC) Health Registry/Cancer Survivorship Cohort (HR/CSC), as well as, the parent study when presenting to the Gynecologic Oncology outpatient clinic from October 2013 to October 2014. Both the HR/CSC and parent study were reviewed and approved by the Human Research Protections Program at the University of North Carolina at Chapel Hill (IRB #09-0605 and #13–2367). Participation included use of medical records for research and interviews. Eligibility criteria included: 18 years or older, North Carolina mailing address, English or Spanish speaking, and planned surgical management of gynecologic cancer or a suspicious pelvic mass.. Exclusion criteria included: pregnancy, the inability to provide informed consent and the inability to participate in an interview. All enrolled patients provided signed informed consent and HIPAA authorization. Participants completed a baseline interview, which included the patient satisfaction questionnaire (PSQ-18). Baseline interviews were conducted within 2 weeks of enrollment by trained staff using a computerassisted telephone interview (CATI) software tool specifically designed for the parent study.
Author Manuscript Author Manuscript
Our primary outcome was patient satisfaction in seven different domains. We measured patient satisfaction using the patient satisfaction questionnaire (PSQ-18). The PSQ-18 is a patient satisfaction tool originally developed by the RAND corporation and has been validated in a diverse number of practice settings (11, 12). It measures patient satisfaction in seven different domains: general satisfaction, satisfaction with the technical quality of care, satisfaction with the interpersonal manner of the physician, satisfaction with communication, satisfaction with the financial aspects of care, satisfaction with time spent with doctor and satisfaction with the accessibility/convenience of care. Each of the 18 questions in the PSQ-18 asks the patient to respond on a 5-point Likert scale and each question maps to a specific domain of patient satisfaction. The minimum number of questions that maps to a single domain is 2 and the maximum number is 4. The score for each domain is a mean score of all questions on that domain. A maximum score is 90 and a minimum score is 18. For our analysis, we converted the PSQ-18 score in each domain of patient satisfaction into a binary outcome of ‘satisfied’ versus ‘unsatisfied/equivocal’ based on a numerical cutoff. Scores of greater than 3.5 were grouped as ‘satisfied’ and scores of 3.5 and below were ‘unsatisfied/equivocal’. This cut off was chosen as, on a single item, a score of 3 indicates being equivocal and a score of 4 indicates being partially satisfied and we wanted to ensure that patients who were categorized as ‘satisfied’ in a specific domain had more responses indicating being ‘satisfied’ than ‘equivocal/unsatisfied’. Our factors of interest were non-modifiable patient factors. These included demographic factors such as age, race, insurance, and highest level of education; geographic factors such as the time travelled to obtain care; and medical factors such as comorbidities, body mass index, specific cancer site, and cancer versus benign disease on final pathology. Health insurance and all medical factors were obtained from the patient’s medical record. Age,
Gynecol Oncol. Author manuscript; available in PMC 2017 August 01.
Barber et al.
Page 4
Author Manuscript
race, education and time travelled to care were reported by the patient in the baseline interview.
Author Manuscript
Associations between non-modifiable patient factors and patient satisfaction in each of the seven domains were evaluated using chi square tests and t-tests. A multivariable logistic regression model was used to estimate the strongest predictors of satisfaction in each domain. Models were created by including all non-modifiable patient factors as covariates; non-modifiable factors that remained significant in these models were considered the strongest predictors of satisfaction. As the goal of the models was to evaluate potential predictors of satisfaction in each of the seven domains rather than test a specific hypothesis, models were not reduced. Sensitivity analyses were performed with reduced models to ensure the odds ratio did not change which would indicate over-fitting. All analyses were performed using SPSS version 20.0 (IBM Corp, Armonk, NY). A p value of less than 0.05 was considered significant for all analyses.
RESULTS Of the 281 women who consented for the parent study, we had patient satisfaction data for 208 (74%). Of the 73 patients without data available, 12 were ineligible due to enrollment criteria or patient withdrawal, 38 did not respond to interview requests and 23 did not have medical data abstracted as they only completed a baseline assessment.
Author Manuscript
Our 208 patients had the following demographic characteristics (Table 1). Median age was 58 years (range 22–93) and median PSQ-18 score was 70.5 (range 42–90). Approximately 20% of the patients were racial/ethnic minorities with 78.4% White patients, 16.8% Black patients and 4.8% Asian, Native American, or self-identified as ‘Other’. Education levels were broadly distributed with 25.5% high school graduate or less, 31.7% some college or trade school, and 42.8% with college degree or greater. Most patients had some private insurance, with 7.2% uninsured and 8.7% with Medicare or Medicaid alone.
Author Manuscript
Several non-modifiable factors were associated with different domains of patient satisfaction (Table 2). Satisfaction with the interpersonal manner of the physician differed for patients depending on race, insurance status, highest level of education and time travelled to care. White patients had higher satisfaction with the interpersonal manner of the physician than racial/ethnic minorities (86% v 65%, p=0.002), as did those with insurance compared to the uninsured. Those with a college education were also more likely to be satisfied with the interpersonal manner of the physician compared to those with a high school education or less (88% v 70%, p=0.03). Those who travelled less than 60 minutes to obtain care were also more likely to be satisfied (88% versus 77%, p=0.05). Satisfaction with the accessibility and convenience of care was significantly different for patients based on race, highest level of education, insurance status, and the time they travelled to care. Patients travelling more than 60 minutes to obtain care were less satisfied with the accessibility and convenience of care than those travelling less than 60 minutes (58% v 74%, p=0.01). Those who were uninsured were also less satisfied with the accessibility of care compared to those with some private insurance. Racial and ethnic
Gynecol Oncol. Author manuscript; available in PMC 2017 August 01.
Barber et al.
Page 5
Author Manuscript
minorities and those with a high school education or less were also significantly more likely to be dissatisfied with the accessibility and convenience of care (p=0.03 and p=0.01, respectively). Finally, satisfaction with financial aspects of care was more common among older patients and insured patients. Seventy-one percent of patients greater than 60 years old were satisfied with financial aspects of care compared to only 47% of those less than 60 years old (p
Gynecol Oncol. Author manuscript; available in PMC 2017 August 01. Published in final edited form as: Gynecol Oncol. 2016 August ; 142(2): 299–303. doi:10.1016/j.ygyno.2016.06.009.
Who Presents Satisfied? Non-modifiable Factors Associated with Patient Satisfaction among Gynecologic Oncology Clinic Patients Emma L Barber, MD1,2, Jeannette T Bensen, PhD2,3, Anna Snavely, PhD4, Paola A Gehrig, MD1,2, and Kemi M Doll, MD1,2,5
Author Manuscript
1Division
of Gynecologic Oncology, Department of Obstetrics and Gynecology, University of North Carolina, Chapel Hill, NC
2Lineberger
Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC
3Department
of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC
4PDstat
LLC
5Department
of Health Policy and Management, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC
Abstract Author Manuscript
Objective—To examine associations between non-modifiable patient factors and patient satisfaction (PS) among women presenting to a gynecologic oncology clinic. Methods—This is a cross sectional analysis of patients presenting for surgical management by a gynecologic oncologist at a tertiary care academic medical center. The Patient Satisfaction Questionnaire (PSQ-18) that measures PS in seven domains of health care was administered. Scores were converted to “satisfied” versus “unsatisfied/equivocal”. Demographic and medical factors were obtained from the medical record. Chi-square, t-tests, and multivariable logistic regression were used.
Author Manuscript
Results—208 patients completed the baseline patient satisfaction questionnaire and the median PSQ-18 score was 70.5 (range 42–90). Median age was 58 years (range 22–93). Several non-modifiable factors were associated with PS. White patients had higher interpersonal PS than minorities (86% v 65% p =0.002). The uninsured had lower interpersonal (60% v 86% p =0.003)
CORRESPONDING AUTHOR: Emma Barber, MD, Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, University of North Carolina at Chapel Hill, 103B Physicians’ Office Building, Campus Box #7572, Chapel Hill, NC 27599. Phone: 919-966-1194; Fax: 919-843-5387; [email protected]. Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. PRESENTATIONS: The data contained within this article was presented in abstract form at both the Annual Winter Meeting of the Society of Gynecologic Oncology February, 2016 in Squaw Valley, CA and at the Society for Gynecologic Oncology Annual Meeting March, 2016 in San Diego, CA. CONFLICT OF INTEREST STATEMENT: The authors declare there are no conflicts of interest.
Barber et al.
Page 2
Author Manuscript
and accessibility PS (33% v 67% p =0.03). Increasing education and less time traveled to care were both associated with higher interpersonal (p =0.03, p =0.05) and accessibility PS (p =0.01, p =0.01). There was no association between clinical factors (BMI, comorbidities, cancer) and PS. In multivariable analysis, the strongest predictor of interpersonal PS was white race while the strongest predictors of accessibility PS were time travelled to care and insurance status. Conclusions—Patient satisfaction scores among patients presenting to a gynecologic oncology clinic are associated with non-modifiable demographic, financial and geographic factors. Pay for performance measures that use summed patient satisfaction scores may penalize hospitals for patient-mix driven differences.
INTRODUCTION Author Manuscript
The Affordable Care Act mandates that payments to hospitals depend on providing valuebased care (1). Two percent of total hospital Center for Medicaid and Medicare Services (CMS) reimbursements is calculated using value-based performance scores (2). Within these value-based performance scores, a subjective measure of performance called the ‘patient experience of care domain’, is 30% of the total hospital performance score (2). This domain is measured using a patient satisfaction questionnaire.
Author Manuscript
Given that patient satisfaction is part of how the delivery of value-based care is measured, many investigators have examined the relationship between patient satisfaction and other markers of high quality healthcare with mixed results. In some studies, higher patient satisfaction scores have been correlated with decreased hospital readmissions, decreased length of hospital stay and decreased mortality (3). Others studies, specifically in surgery, have found no relationship between patient satisfaction and adherence to Surgical Care Improvement Program (SCIP) process measures, postoperative complications or mortality (4, 5). Paradoxically, one study actually found an inverse relationship: patient satisfaction was associated with increased healthcare costs and increased mortality (6).
Author Manuscript
In addition to studying the relationship between patient satisfaction and other markers of high quality healthcare, investigators have also examined the association between nonmodifiable patient factors and patient satisfaction. In this work, patient satisfaction scores vary with a number of non-modifiable patient and hospital system factors. At the patient level, lower satisfaction scores have been found for younger and more educated patients (7, 8). Other studies have found lower satisfaction scores for patients who are non-English speakers (8). At the systems level, lower patient satisfaction scores have been observed for densely populated urban areas, hospitals with more beds and those that are non-profit or academic (8, 9). The consistent variation of patient satisfaction by non-modifiable factors suggests that patient satisfaction may be measuring factors beyond just the quality of healthcare delivered. However, much of this literature focuses on the inpatient setting, and the association between non-modifiable factors and patient satisfaction has not been studied to date in outpatient gynecologic oncology patients. Our objective was to evaluate the association between non-modifiable patient factors and domain-specific patient satisfaction scores among women presenting to a gynecologic oncology clinic. Gynecol Oncol. Author manuscript; available in PMC 2017 August 01.
Barber et al.
Page 3
Author Manuscript
METHODS
Author Manuscript
We conducted a cross sectional analysis of a larger parent prospective cohort study: The Impact of Surgical Complications on Health Related Quality of Life.(10) The recruitment of this cohort has been previously described. Briefly, patients were identified and enrolled in a hospital-based Registry, the University of North Carolina at Chapel Hill (UNC) Health Registry/Cancer Survivorship Cohort (HR/CSC), as well as, the parent study when presenting to the Gynecologic Oncology outpatient clinic from October 2013 to October 2014. Both the HR/CSC and parent study were reviewed and approved by the Human Research Protections Program at the University of North Carolina at Chapel Hill (IRB #09-0605 and #13–2367). Participation included use of medical records for research and interviews. Eligibility criteria included: 18 years or older, North Carolina mailing address, English or Spanish speaking, and planned surgical management of gynecologic cancer or a suspicious pelvic mass.. Exclusion criteria included: pregnancy, the inability to provide informed consent and the inability to participate in an interview. All enrolled patients provided signed informed consent and HIPAA authorization. Participants completed a baseline interview, which included the patient satisfaction questionnaire (PSQ-18). Baseline interviews were conducted within 2 weeks of enrollment by trained staff using a computerassisted telephone interview (CATI) software tool specifically designed for the parent study.
Author Manuscript Author Manuscript
Our primary outcome was patient satisfaction in seven different domains. We measured patient satisfaction using the patient satisfaction questionnaire (PSQ-18). The PSQ-18 is a patient satisfaction tool originally developed by the RAND corporation and has been validated in a diverse number of practice settings (11, 12). It measures patient satisfaction in seven different domains: general satisfaction, satisfaction with the technical quality of care, satisfaction with the interpersonal manner of the physician, satisfaction with communication, satisfaction with the financial aspects of care, satisfaction with time spent with doctor and satisfaction with the accessibility/convenience of care. Each of the 18 questions in the PSQ-18 asks the patient to respond on a 5-point Likert scale and each question maps to a specific domain of patient satisfaction. The minimum number of questions that maps to a single domain is 2 and the maximum number is 4. The score for each domain is a mean score of all questions on that domain. A maximum score is 90 and a minimum score is 18. For our analysis, we converted the PSQ-18 score in each domain of patient satisfaction into a binary outcome of ‘satisfied’ versus ‘unsatisfied/equivocal’ based on a numerical cutoff. Scores of greater than 3.5 were grouped as ‘satisfied’ and scores of 3.5 and below were ‘unsatisfied/equivocal’. This cut off was chosen as, on a single item, a score of 3 indicates being equivocal and a score of 4 indicates being partially satisfied and we wanted to ensure that patients who were categorized as ‘satisfied’ in a specific domain had more responses indicating being ‘satisfied’ than ‘equivocal/unsatisfied’. Our factors of interest were non-modifiable patient factors. These included demographic factors such as age, race, insurance, and highest level of education; geographic factors such as the time travelled to obtain care; and medical factors such as comorbidities, body mass index, specific cancer site, and cancer versus benign disease on final pathology. Health insurance and all medical factors were obtained from the patient’s medical record. Age,
Gynecol Oncol. Author manuscript; available in PMC 2017 August 01.
Barber et al.
Page 4
Author Manuscript
race, education and time travelled to care were reported by the patient in the baseline interview.
Author Manuscript
Associations between non-modifiable patient factors and patient satisfaction in each of the seven domains were evaluated using chi square tests and t-tests. A multivariable logistic regression model was used to estimate the strongest predictors of satisfaction in each domain. Models were created by including all non-modifiable patient factors as covariates; non-modifiable factors that remained significant in these models were considered the strongest predictors of satisfaction. As the goal of the models was to evaluate potential predictors of satisfaction in each of the seven domains rather than test a specific hypothesis, models were not reduced. Sensitivity analyses were performed with reduced models to ensure the odds ratio did not change which would indicate over-fitting. All analyses were performed using SPSS version 20.0 (IBM Corp, Armonk, NY). A p value of less than 0.05 was considered significant for all analyses.
RESULTS Of the 281 women who consented for the parent study, we had patient satisfaction data for 208 (74%). Of the 73 patients without data available, 12 were ineligible due to enrollment criteria or patient withdrawal, 38 did not respond to interview requests and 23 did not have medical data abstracted as they only completed a baseline assessment.
Author Manuscript
Our 208 patients had the following demographic characteristics (Table 1). Median age was 58 years (range 22–93) and median PSQ-18 score was 70.5 (range 42–90). Approximately 20% of the patients were racial/ethnic minorities with 78.4% White patients, 16.8% Black patients and 4.8% Asian, Native American, or self-identified as ‘Other’. Education levels were broadly distributed with 25.5% high school graduate or less, 31.7% some college or trade school, and 42.8% with college degree or greater. Most patients had some private insurance, with 7.2% uninsured and 8.7% with Medicare or Medicaid alone.
Author Manuscript
Several non-modifiable factors were associated with different domains of patient satisfaction (Table 2). Satisfaction with the interpersonal manner of the physician differed for patients depending on race, insurance status, highest level of education and time travelled to care. White patients had higher satisfaction with the interpersonal manner of the physician than racial/ethnic minorities (86% v 65%, p=0.002), as did those with insurance compared to the uninsured. Those with a college education were also more likely to be satisfied with the interpersonal manner of the physician compared to those with a high school education or less (88% v 70%, p=0.03). Those who travelled less than 60 minutes to obtain care were also more likely to be satisfied (88% versus 77%, p=0.05). Satisfaction with the accessibility and convenience of care was significantly different for patients based on race, highest level of education, insurance status, and the time they travelled to care. Patients travelling more than 60 minutes to obtain care were less satisfied with the accessibility and convenience of care than those travelling less than 60 minutes (58% v 74%, p=0.01). Those who were uninsured were also less satisfied with the accessibility of care compared to those with some private insurance. Racial and ethnic
Gynecol Oncol. Author manuscript; available in PMC 2017 August 01.
Barber et al.
Page 5
Author Manuscript
minorities and those with a high school education or less were also significantly more likely to be dissatisfied with the accessibility and convenience of care (p=0.03 and p=0.01, respectively). Finally, satisfaction with financial aspects of care was more common among older patients and insured patients. Seventy-one percent of patients greater than 60 years old were satisfied with financial aspects of care compared to only 47% of those less than 60 years old (p
